This is a new R01 proposal addressing neural pathways and gene expression in habenula- midbrain circuits. The habenula is a dorsal thalamic nucleus consisting of medial (MH) and lateral (LH) subnuclei that participate in distinct neural pathways. The MH and LH both send their output fibers to the ventral midbrain via the prominent fasciculus retroflexus (FR), but only the LH directly innervates midbrain dopamine (DA) and serotonin (5HT) systems, while the MH projects first to the interpeduncular nucleus (IP), which in turn projects to ventral midbrain. Recent behavioral experiments suggest strong functional interactions between the habenula-midbrain pathway and the monoamine systems, with major implications for mood disorders, cognitive disorders, and addiction. Using microarrays and bioinformatic approaches, we have recently shown that molecular markers identify subpopulations of neurons in the MH and LH which suggest functional heterogeneity within these nuclei. Further progress in understanding habenula function will require a much better understanding of these distinct molecular classes of habenula neurons, including their anatomic and functional connectivity, which are not accessible to conventional tract-tracing methods. Here we propose three Specific Aims that will explore habenula circuits and rapidly delineate the connections of defined populations of neurons in the MH, LH and IP. These experiments are planned in a short time-frame (3 years) and with cost-effective use of existing tools, including genomic and bioinformatic assets from the Gensat and Allen Brain Institute brain mapping projects. Aim 1. Test the hypothesis that specific subpopulations of MH and LH neurons participate in distinct neural pathways. Use transgenic mice with LacZ, GFP, and Cre reporters to trace the projections of genetically defined classes of MH, LH, and IP neurons to the midbrain. Available lines target the Brn3a, Gpr151, Tac2/SK, Slc18a3, Chrnb4, Prokr2, Chrna5, Tac1/SP, and Chat genes. Aim 2. Test the hypothesis that the LH provides input to the mesolimbic DA system via GABAergic neurons of the mesopontine rostromedial tegmental nucleus (RMTg) in mice as it does in rats. Localize the mouse RMTg using cFos induction, anterograde tract tracing from the LH, and GABA marker expression. Identify the LH subpopulations projecting to the RMTg using transgenic tracing. Aim 3. In collaboration with the Allen Institute, use bioinformatic analysis to test the hypothesis that key nodes of the habenula pathway, specifically the forebrain septal nuclei, LH, IP and RMTg, have distinctive gene expression profiles, as previously demonstrated for the MH. These gene expression profiles will form the basis for the genetic manipulation of habenula pathways in future studies.